In global science and remote sensing programmes there is a need for technological, institutional and intellectual resources to store, conceptualise, process and visualise the data coming in. There is a real data assimilation problem, which has to deal with errors and uncertainties as well as parameterisation and scaling issues. What are required are sources of data about the present status of resources and trends over time, conceptual models and prediction engines to assimilate the data and turn it into information, and institutions and systems to enable action to be taken where required. With the explosion of data and information systems in the past two or three decades, it is the institutional and governance systems that we are lacking the most. Data systems provide information, institutional and governance systems allow management action to be taken, but it is values and beliefs that ultimately determine whether anything is done.

In global meteorological observation and weather forecasting we now have some very sophisticated systems to receive the satellite observations as well as predictive models to assimilate the data as they are received. Models of the global atmospheric circulation are continuously updated by streams of detailed information about the present state of the atmosphere. Huge investments have been made in solving some of the problems of data fusion and assimilation across scales and between image and point source data.